Plants may contribute to the biodegradation of polycyclic aromatic hydrocarbons (PAHs) in contaminated soils. Different mechanisms have been proposed, such as an increase in microbial numbers, but are not clearly elucidated. This study investigates the dissipation of a mixture of eight PAHs, ranging from 3 to 6 rings, in the rhizosphere of ryegrass ( Lolium perenne L.). Two pot experiments were conducted with or without plants using soil spiked with 1 g kg −1 of PAHs in a growth chamber. The first experiment was carried out shortly after spiking and the second after 6 months of ageing. At the end of both experiments, the extractable concentrations of all PAHs were lower in rhizospheric than non-rhizospheric soil. PAH dissipation was lower after soil ageing than before, but was still significantly higher in the rhizospheric soil, even for three of the high molecular weight PAHs. Total culturable microflora were higher in the rhizospheric than non-rhizospheric soil, but was at the same level in spiked and non-spiked soil. The number of PAHs degraders, estimated by a modified MPN procedure, was not significantly different in the freshly spiked rhizospheric and non-rhizospheric soils, but was significantly higher in the rhizosphere of the aged spiked soil.